CN110309878A - A kind of information processing method and device, storage medium - Google Patents
A kind of information processing method and device, storage medium Download PDFInfo
- Publication number
- CN110309878A CN110309878A CN201910580391.1A CN201910580391A CN110309878A CN 110309878 A CN110309878 A CN 110309878A CN 201910580391 A CN201910580391 A CN 201910580391A CN 110309878 A CN110309878 A CN 110309878A
- Authority
- CN
- China
- Prior art keywords
- information
- class
- time
- trip
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 96
- 238000003672 processing method Methods 0.000 title claims abstract description 19
- 238000000034 method Methods 0.000 claims abstract description 35
- 230000004044 response Effects 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims description 36
- 230000015654 memory Effects 0.000 claims description 22
- 238000004422 calculation algorithm Methods 0.000 claims description 14
- 238000012512 characterization method Methods 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 6
- 241000406668 Loxodonta cyclotis Species 0.000 claims description 3
- 238000012216 screening Methods 0.000 description 27
- 230000001815 facial effect Effects 0.000 description 25
- 238000010586 diagram Methods 0.000 description 12
- 238000012544 monitoring process Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000011835 investigation Methods 0.000 description 2
- 241000251468 Actinopterygii Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006116 polymerization reaction Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Tourism & Hospitality (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Resources & Organizations (AREA)
- Educational Administration (AREA)
- Probability & Statistics with Applications (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Alarm Systems (AREA)
Abstract
Present embodiment discloses a kind of information processing method and devices, storage medium, this method may include: receive first appear inquiry instruction in the case where, in response to first appearing inquiry instruction, target area, object time and target duration are obtained in inquiry instruction from first appearing;From presetting at least one object trip information, at least one the candidate target trip information of trip information in target area is searched;From at least one candidate target trip information, search that not occurring in the target duration before object time and what is occurred after object time first appear object information;Display first appears object information.Embodiment improves search the feasibility and intelligence for first appearing object.
Description
Technical field
This disclosure relates to computer vision field more particularly to a kind of information processing method and device, storage medium.
Background technique
Currently, public security department is when carrying out investigations to case or deploying to ensure effective monitoring and control of illegal activities to specific region, due to first appearing pair
The crime possibility of elephant is larger, will often first appear object as the investigation target of case or the object of deploying to ensure effective monitoring and control of illegal activities of specific region,
It first appears object and just refers to and do not appear in target area within a period of time before public security department organizes work, but opening up
It goes into operation and appears in the object of target area after making.
In general, by police strength personnel target area is monitored in real time or to the moving object in current region
It is visited and investigated, first appears object to determine, however, this artificial lookup first appears the feasibility and intelligence of object
It is low.
Summary of the invention
The present embodiment proposes a kind of information processing method and device, storage medium, can be improved lookup and first appears object
Feasibility and intelligence.
The technical solution of the disclosure is achieved in that
Present embodiments provide a kind of information processing method, which comprises
Receive first appear inquiry instruction in the case where, inquiry instruction is first appeared in response to described, from the head
It is secondary to occur obtaining target area, object time and target duration in inquiry instruction;
From presetting at least one object trip information, at least one time of trip information in the target area is searched
Select object trip information;
From at least one described candidate target trip information, the target duration before the object time is searched
It is not occurring inside and what is occurred after the object time first appears object information;
Object information is first appeared described in display.
In above scheme, described to preset at least one object trip information each object trip information include line position
It sets;It is described from presetting at least one object trip information, search at least one time of trip information in the target area
Select object trip information, comprising:
The trip information that trip position in each object trip information is not belonging to the target area is deleted, is obtained
To at least one described candidate target trip information.
In above scheme, each candidate target trip information includes each at least one described candidate target trip information
The travel time of candidate target;It is described from least one described candidate target trip information, search the object time it
It is occurring in the preceding target duration and what is occurred after the object time first appears object information, comprising:
According to the travel time of the object time and each candidate target, whether each candidate target is judged
For the object occurred after the object time;
It is in the case where the object occurred after the object time, to each time in each candidate target
The travel time of object is selected to be counted, when determining that history of each candidate target before the object time does not occur
It is long;
In the case where the history does not occur duration not less than the target duration, each candidate target is gone on a journey
Information first appears object information described in being used as.
In above scheme, described from presetting at least one object trip information, trip information is searched in the target
Before at least one candidate target trip information in region, the method also includes:
Obtain target case information;
According to the target case information, determine that at least one is initial at least one object trip information from described preset
Object trip information;
Correspondingly, described from presetting at least one object trip information, trip information is searched in the target area
At least one candidate target trip information, comprising:
From at least one described initial object trip information, at least one of trip information in the target area is searched
A candidate target trip information.
In above scheme, described from presetting at least one object trip information, trip information is searched in the target
Before at least one candidate target trip information in region, the method also includes:
Obtain object screening instruction;
It is screened and is instructed according to the object, determine that at least one is initial at least one object trip information from described preset
Object trip information;
Correspondingly, described from presetting at least one object trip information, trip information is searched in the target area
At least one candidate target trip information, comprising:
From at least one described initial object trip information, at least one of trip information in the target area is searched
A candidate target trip information.
In above scheme, the object screening instruction includes that the first object screens instruction and/or the screening instruction of the second object,
The first object screening instruction is for selecting the first class object, and the second object screening instruction is for excluding the second class pair
As.
In above scheme, described from least one described candidate target trip information, search in the object time
It is not occurring in the target duration before and after what is occurred after the object time first appear object information, institute
It states before first appearing object information described in display, the method also includes:
It is first appeared in object information from described, obtains at least one and first appear and each first appear object in object
At least one travel time and at least one trip position;
Obtain current time;
According to preset time difference computation model, at least one described travel time and the current time, obtain it is described extremely
Few at least one corresponding time difference of the travel time;Preset time difference computation model characterization input value and output valve
Corresponding relationship, the input value are travel time and current time, and the output valve is the time difference;
According at least one described time difference, described each first appear pair is determined from least one described travel time
The newest time of occurrence of elephant and from it is described at least one trip position in determine it is described each first appear object it is newest go out
Existing position;
Correspondingly, object information is first appeared described in the display, comprising:
Display it is described at least one first appear object, the newest time of occurrence for each first appearing object and
The newest appearance position for each first appearing object.
In above scheme, inquiry instruction is first appeared in response to described described, is first appeared in inquiry instruction from described
Before obtaining target area, object time and target duration, the method also includes:
In the case where reaching at current preset cluster time point, previous default cluster time point and the current preset are obtained
It clusters acquired image between time point and acquires information;
According to default clustering algorithm, described image acquisition information and default identity library, obtain that described to preset at least one right
As trip information.
In above scheme, it includes acquisition image and acquisition space time information that described image, which acquires information,;The basis is default poly-
Class algorithm, described image acquisition information and default identity library obtain described presetting at least one object trip information, comprising:
An at least identity characteristic image is obtained from the acquisition image;
Using the default clustering algorithm, an at least identity characteristic image is clustered, at least one is obtained
Feature class;The first similarity at least one described feature class between the corresponding identity characteristic image of each feature class is greater than pre-
If the first similarity threshold;
At least one corresponding trip information of feature class is determined from the acquisition space time information;
According to the default identity library and at least one described feature class, it is corresponding extremely to obtain at least one fisrt feature class
Few class object information;At least one described fisrt feature class belongs at least one described feature class, the default identity library
It is stored at least one default object information, at least one described class object information belongs at least one described default object letter
Breath;
According at least one described corresponding trip information of feature class and at least one described fisrt feature class pair
At least one the class object information answered obtains described presetting at least one object trip information.
It is described according to the default identity library and at least one described feature class in above scheme, obtain at least one
At least one corresponding class object information of one feature class, comprising:
At least one corresponding class central feature value of at least one described feature class is obtained respectively;
Calculate one group of at least one described default object information and the corresponding current class central feature value of current signature class
Second similarity, the corresponding current class central feature value of the current signature class are at least one described class central feature value
Any one class central feature value;
Target similarity is determined from one group of second similarity;
In the case where the target similarity is not less than default second similarity threshold, using the current signature class as
Fisrt feature class, and from the default identity library, determine the corresponding default object information of the target similarity value;
The corresponding default object information of the target similarity value is determined as the corresponding class object of the fisrt feature class
Information, and then obtain at least one corresponding class object information of at least one described fisrt feature class.
In above scheme, described at least one corresponding trip information of feature class according to and it is described at least
At least one corresponding class object information of one fisrt feature class obtains described presetting at least one object trip information, comprising:
From at least one described corresponding trip information of feature class, determine that at least one described fisrt feature class is each
Self-corresponding first trip information and corresponding second trip information of at least one second feature class, described second is special
Sign class belongs at least one described feature class;
It will corresponding first trip information of at least one fisrt feature class and at least one described first spy
At least one corresponding class object information of sign class is mapped, and real name trip information is obtained;
By corresponding second trip information of at least one described second feature class, as non-real name trip information;
By the real name trip information and the non-real name trip information, at least one object trip letter is preset as described
Breath.
In above scheme, described at least one corresponding trip information of feature class according to and it is described at least
At least one corresponding class object information of one fisrt feature class, obtain it is described preset at least one object trip information after,
The method also includes:
In the case where getting new default object information, by the new default object information and the default identity
The default object information polymerization in library, obtains default object information to be updated;
According to the default object information to be updated, update is described to preset at least one object trip information, is updated
Object trip information afterwards;
Correspondingly, described from presetting at least one object trip information, trip information is searched in the target area
At least one candidate target trip information, comprising:
From the updated object trip information, at least one time of trip information in the target area is searched
Select object trip information.
In above scheme, described at least one corresponding trip information of feature class according to and it is described at least
At least one corresponding class object information of one fisrt feature class, obtain it is described preset at least one object trip information after,
The method also includes:
In the case where reaching at next default cluster time point, current preset cluster time point and described next default is obtained
It clusters collected new images between time point and acquires information;
At least one new identity characteristic image is obtained from new images acquisition information;
According at least one described feature class, increment cluster meter is carried out to described at least one new identity characteristic image
It calculates, obtains at least one feature class to be added;
From new images acquisition information, at least one described feature class to be added of acquisition is corresponding respectively newly goes out
Row information;
According at least one described corresponding new trip information of feature class to be added, update is described to preset at least one
Object trip information obtains updated object trip information;
Correspondingly, described from presetting at least one object trip information, trip information is searched in the target area
At least one candidate target trip information, comprising:
From the updated object trip information, at least one time of trip information in the target area is searched
Select object trip information.
A kind of information processing unit is present embodiments provided, described device includes:
Instruction receiving unit, for receive first appear inquiry instruction in the case where, first appeared in response to described
Inquiry instruction obtains target area, object time and target duration from described first appear in inquiry instruction;
First processing units, for searching trip information in the target from presetting at least one object trip information
At least one candidate target trip information in region;
The second processing unit, for searching in the object time from least one described candidate target trip information
It is not occurring in the target duration before and what is occurred after the object time first appears object information;
Display unit described first appears object information for showing.
In above scheme, described to preset at least one object trip information each object trip information include line position
It sets;The first processing units, specifically for the trip position in each object trip information is not belonging to the target
The trip information in region is deleted, at least one described candidate target trip information is obtained.
In above scheme, each candidate target trip information includes each at least one described candidate target trip information
The travel time of candidate target;
Described the second processing unit, when specifically for according to the trip of the object time and each candidate target
Between, judge whether each candidate target is the object occurred after the object time;And it is right in each candidate
As to be united in the case where the object occurred after the object time to the travel time of each candidate target
Meter, determines that history of each candidate target before the object time does not occur duration;And do not go out in the history
It is current long not less than in the case where the target duration, each candidate target trip information is first appeared pair as described in
Image information.
In above scheme, the first processing units are also used to described from presetting at least one object trip information,
Before searching at least one the candidate target trip information of trip information in the target area, target case information is obtained;
According to the target case information, the trip of at least one initial object is determined at least one object trip information from described preset
Information;
Correspondingly, the first processing units are specifically used for from least one described initial object trip information, search
At least one the candidate target trip information of trip information in the target area.
In above scheme, the first processing units are also used to described from presetting at least one object trip information,
Before searching at least one the candidate target trip information of trip information in the target area, object screening instruction is obtained;
It is screened and is instructed according to the object, determine the trip of at least one initial object at least one object trip information from described preset
Information;
Correspondingly, the first processing units are specifically used for from least one described initial object trip information, search
At least one the candidate target trip information of trip information in the target area.
In above scheme, the object screening instruction includes that the first object screens instruction and/or the screening instruction of the second object,
The first object screening instruction is for selecting the first class object, and the second object screening instruction is for excluding the second class pair
As.
In above scheme, described the second processing unit is also used to believe in described go on a journey from least one described candidate target
In breath, searches not occurring in the target duration before the object time and occur after the object time
After first appearing object information, before first appearing object information described in the display, object information is first appeared from described
In, it obtains at least one and first appears at least one travel time for each first appearing object in object and at least one trip
Position;And obtain current time;And according to preset time difference computation model, at least one described travel time and described current
Moment obtains described at least one corresponding time difference of at least one travel time;The preset time difference computation model characterization
The corresponding relationship of input value and output valve, the input value are travel time and current time, and the output valve is the time difference;With
And according at least one described time difference, determination is described from least one described travel time each first appears object most
New time of occurrence and the determining newest appearance position for each first appearing object from least one described trip position
It sets;
Correspondingly, the display unit, specifically for show it is described at least one first appear object, it is described it is each for the first time
There is the newest time of occurrence of object and the newest appearance position for each first appearing object.
In above scheme, described device further include:
Information acquisition unit first appears inquiry from described for first appearing inquiry instruction in response to described described
Before obtaining target area, object time and target duration in instruction, in the case where reaching at current preset cluster time point, obtain
Previous default cluster time point and the current preset is taken to cluster acquired image acquisition information between time point;And according to
Default clustering algorithm, described image acquisition information and default identity library obtain described presetting at least one object trip information.
In above scheme, it includes acquisition image and acquisition space time information that described image, which acquires information,;The acquisition of information list
Member, specifically for obtaining an at least identity characteristic image from the acquisition image;And the default clustering algorithm is utilized, it is right
An at least identity characteristic image is clustered, at least one feature class is obtained;It is each at least one described feature class
The first similarity between the corresponding identity characteristic image of feature class is greater than default first similarity threshold;And from when the acquisition
At least one corresponding trip information of feature class is determined in empty information;And according to the default identity library and described at least one
A feature class obtains at least one corresponding class object information of at least one fisrt feature class;At least one described fisrt feature
Class belongs at least one described feature class, and the default identity inventory contains at least one default object information, and described at least one
A class object information belongs at least one described default object information;And it is corresponding according at least one described feature class
Trip information and at least one corresponding class object information of at least one described fisrt feature class obtain described preset at least
One object trip information.
In above scheme, it is corresponding to be specifically used at least one described feature class of acquisition respectively for the information acquisition unit
At least one class central feature value;And it calculates in described at least one default object information and the corresponding current class of current signature class
One group of second similarity of heart characteristic value, the corresponding current class central feature value of the current signature class are at least one described class
Any one class central feature value in central feature value;And target similarity is determined from one group of second similarity;And
In the case that the target similarity is not less than default second similarity threshold, using the current signature class as fisrt feature
Class, and from the default identity library, determine the corresponding default object information of the target similarity value;And by the target
The corresponding default object information of similarity value is determined as the corresponding class object information of the fisrt feature class, so obtain it is described extremely
Few at least one corresponding class object information of a fisrt feature class.
In above scheme, the information acquisition unit is specifically used for corresponding out from least one described feature class
In row information, corresponding first trip information of at least one fisrt feature class and at least one second spy are determined
Corresponding second trip information of class is levied, the second feature class belongs at least one described feature class;And by described at least
Corresponding first trip information of one fisrt feature class and at least one described fisrt feature class it is corresponding at least one
Class object information is mapped, and real name trip information is obtained;And it will at least one described second feature class corresponding second
Trip information, as non-real name trip information;And by the real name trip information and the non-real name trip information, as institute
It states and presets at least one object trip information.
In above scheme, described device further include: first information updating unit, for it is described according at least one
The corresponding trip information of feature class and at least one corresponding class object information of at least one described fisrt feature class,
Obtain it is described preset at least one object trip information after, will be described in the case where getting new default object information
New default object information polymerize with the default object information in the default identity library, obtains default object letter to be updated
Breath;And according to the default object information to be updated, update is described to preset at least one object trip information, obtains updated
Object trip information;
Correspondingly, the first processing units are specifically used for from the updated object trip information, search trip
At least one the candidate target trip information of information in the target area.
In above scheme, described device further include: the second information updating unit, for it is described according at least one
The corresponding trip information of feature class and at least one corresponding class object information of at least one described fisrt feature class,
Obtain it is described preset at least one object trip information after, next default cluster time point reach in the case where, acquisition is worked as
Collected new images acquire information between preceding default cluster time point and next default cluster time point;And from described new
At least one new identity characteristic image is obtained in Image Acquisition information;And according at least one described feature class, to it is described extremely
A few new identity characteristic image carries out increment cluster calculation, obtains at least one feature class to be added;And from the new figure
As obtaining at least one described corresponding new trip information of feature class to be added respectively in acquisition information;And according to institute
At least one corresponding new trip information of feature class to be added is stated, update is described to preset at least one object trip information,
Obtain updated object trip information;
Correspondingly, the first processing units are specifically used for from the updated object trip information, search trip
At least one the candidate target trip information of information in the target area.
Present embodiments providing a kind of information processing unit, described device includes: processor, memory and communication bus,
The memory is communicated by the communication bus with the processor, and it is executable that the memory stores the processor
One or more program pass through the processor and execute such as above-mentioned when one or more of programs are performed
The step of one information processing method.
A kind of computer readable storage medium is present embodiments provided, the computer-readable recording medium storage has journey
Sequence causes at least one described processor to execute any of the above-described such as and believes when described program is executed by least one processor
The step of ceasing processing method.
The present embodiment provides a kind of information processing method and device, storage medium, using above-mentioned technic relization scheme, due to
The trip information that at least one object trip information characterizes object is preset, therefore, information processing unit is looked into according to first appearing
Target area, object time and the target duration in instruction are ask, from presetting at least one object trip information, it will be able to obtain
It is in the target area, not occurring in the target duration before the object time to trip information and
What is occurred after the object time first appears object information, and then also can be quick, intelligently based on object of going on a journey for the first time
Information determines to first appear object, and the present embodiment improves the feasibility and intelligence searched and first appear object.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and those figures show meet this public affairs
The embodiment opened, and together with specification it is used to illustrate the technical solution of the disclosure.
Fig. 1 is a kind of implementation flow chart one of information processing method provided in this embodiment;
Fig. 2 is a kind of page schematic diagram of illustrative information processing method provided in this embodiment;
Fig. 3 is a kind of implementation flow chart two of information processing method provided in this embodiment;
Fig. 4 is a kind of implementation flow chart for obtaining object trip information provided in this embodiment;
Fig. 5 is that a kind of illustrative facial biometric provided in this embodiment clusters schematic diagram;
Fig. 6 is the flow diagram that one kind provided in this embodiment illustratively obtains at least one feature class;
Fig. 7 is a kind of implementation flow chart of illustrative increment cluster provided in this embodiment;
Fig. 8 is a kind of flow diagram of illustrative upgating object trip information provided in this embodiment;
Fig. 9 is a kind of structural schematic diagram one of information processing unit provided in this embodiment;
Figure 10 is a kind of structural schematic diagram two of information processing unit provided in this embodiment.
Specific embodiment
It should be appreciated that specific embodiment described herein is only used to explain the disclosure.It is not used to limit the disclosure.
Embodiment one
A kind of information processing method is present embodiments provided, Fig. 1 is that a kind of information processing method provided in this embodiment is real
Existing flow chart, as shown in Figure 1, the information processing method includes:
S101, receive first appear inquiry instruction in the case where, in response to first appearing inquiry instruction, go out from for the first time
Target area, object time and target duration are obtained in existing inquiry instruction;
Information processing unit receive user sending first appear inquiry instruction, obtained from first appearing in inquiry instruction
Target area, object time and target duration, with according to target area, object time and target duration, from presetting at least one
It determines to first appear object in object trip information;Wherein, preset at least one object trip information characterization trip information and
The corresponding relationship of class object information, class object information be the relevant information of citizenship, for example, name, gender, identification card number and
The information such as social action information.
It should be noted that target area includes at least one location information, for example, can be one or more cameras
Or the corresponding longitude and latitude of capture machine, it is also possible to one or more regions;Object time is any one historical juncture, for example,
Moment, start time of deploying to ensure effective monitoring and control of illegal activities etc. occur for case, the duration before object time that target duration refers to, can for a week or
One month etc..
In the present embodiment, target case information, target case information can also be obtained in inquiry instruction from first appearing
Refer to the target object information of case to be checked, for example, target object gender, target object criminal type etc..
Illustratively, for including that target area, object time and target duration first appear inquiry instruction, at information
Device is managed from least one corresponding object of feature class, determination does not appear in target in the target duration before object time
Target area is appeared in region and after object time first appears object.
Illustratively, for going out for the first time including target area, object time, target duration and target object criminal type
Existing inquiry instruction, due to presetting the corresponding relationship of at least one object trip information characterization trip information and class object information, such as
Fruit object information includes criminal type, then, information processing unit can be from presetting at least one object trip information, really
Do not appeared in target area in target duration before being scheduled on object time, appeared in after object time target area,
And class object information includes that target object criminal type first appears object, will first appear object and sends out as object time
The suspicion personnel of the raw case to be checked in target area, the criminal type of case to be checked are target object criminal type.
S102, from presetting at least one object trip information, search at least one of trip information in target area
Candidate target trip information;
Information processing unit obtains trip position and belongs in target area from presetting at least one object trip information
At least one candidate target trip information.
In the present embodiment, presetting each object trip information at least one object trip information includes trip position;
Information processing unit by the trip position in each object trip information be not belonging to target area trip information delete, obtain to
A few candidate target trip information.
In the present embodiment, information processing unit searches trip information from presetting at least one object trip information
Before at least one candidate target trip information in target area, target case information is obtained;According to target case information,
At least one initial object trip information is determined at least one object trip information from presetting;Correspondingly, information processing unit
It can search at least one candidate target of trip information in target area from least one initial object trip information and go out
Row information.
Information processing unit can also obtain target case information from first appearing in inquiry instruction, go on a journey from each object
Class object information is obtained in information;Judge class object information in target case information and each object trip information whether information
Unanimously, when information is consistent, using each object trip information as initial object trip information, otherwise each object trip information
It cannot function as initial object trip information, and then obtain at least one initial object trip information;It is initial right from least one again
As searching at least one candidate target trip information in trip information.
It should be noted that searching at least one candidate target trip information from least one initial object trip information
Process, and from the process for searching at least one candidate target trip information at least one object trip information is preset, similarly.
Illustratively, by taking target case information includes target object gender and target object criminal type as an example, class object
Information includes gender and criminal type etc., in the gender in class object information as target object gender and class object information
In criminal type as target criminal type in the case where, determine that class object information is consistent with target case information;In class
Gender and target object gender in object information be different, and/or criminal type and target crime class in class object information
In the different situation of type, determine that class object information and target case information are inconsistent.
In the present embodiment, from presetting at least one object trip information, trip information is searched in target area
At least one candidate target trip information before, information processing unit obtain object screening instruction;It is screened and is instructed according to object,
At least one initial object trip information is determined at least one object trip information from presetting;Correspondingly, from default at least one
In a object trip information, search at least one candidate target trip information in target area of trip information, comprising: to
In a few initial object trip information, at least one the candidate target trip information of trip information in target area is searched.
Information processing unit obtains object screening instruction, object screening instruction characterization object filter information;It is sieved according to object
Information is selected, determines at least one initial object trip information at least one object trip information from presetting;Again from least one
In initial object trip information, at least one the candidate target trip information of trip information in target area is searched.
It should be noted that information processing unit can receive object sieve while reception first appears inquiry instruction
Choosing instruction.
In the present embodiment, object screening instruction includes that the first object screens instruction and/or the screening instruction of the second object, the
An object screening instruction is for selecting the first class object, and the screening instruction of the second object is for excluding the second class object.
In the present embodiment, the first object screening instruction includes selection identity library, and the first class object refers to selection identity library
In object, select identity library include the first class object identity information;Information processing unit judges each object trip information
In class object information whether belong to selection identity library, if belonged to, using each object trip information as initial object trip
Information otherwise by each object trip information not as initial object trip information, and then obtains at least one initial object and goes out
Row information.
In the present embodiment, the second object screening instruction includes excluding identity library, and the second class object, which refers to, excludes identity library
In object, exclude identity library include the second class object identity information;Information processing unit judges each object trip information
In class object information whether belong to selection identity library, if belonged to, using each object trip information as initial object trip
Information otherwise by each object trip information not as initial object trip information, and then obtains at least one initial object and goes out
Row information.
In the present embodiment, information processing unit searches trip information from presetting at least one object trip information
Before at least one candidate target trip information in target area, target case information and object screening instruction are obtained;Root
According to target case information and object screening instruction, at least one initial object is determined at least one object trip information from presetting
Trip information.
Information processing unit selects first according to target case information at least one object trip information is preset, then
It is screened and is instructed according to object, at least one initial object trip information is determined from the object trip information after selection;Alternatively, first
It is screened and is instructed according to object, selected at least one object trip information is preset, further according to target case information, from selection
At least one initial object trip information is determined in object trip information afterwards.
S103, from least one candidate target trip information, lookup do not go out in the target duration before object time
It is existing and what is occurred after object time first appears object information;
Information processing unit meets the trip information at least one candidate target trip information before object time
Target duration in do not occur and occur after object time, as first appearing object information.
In the present embodiment, each candidate target trip information includes each time at least one candidate target trip information
Select the travel time of object;Information processing unit judges each time according to the travel time of object time and each candidate target
Select whether object is the object occurred after object time;It is the object occurred after object time in each candidate target
In the case where, the travel time of each candidate target is counted, determines each candidate target going through before object time
There is duration in Shi Wei;In the case where history does not occur duration not less than target duration, each candidate target trip information is made
To first appear object information.
Information processing unit does not include the feelings of the time after object time in the travel time of a candidate target
Under condition, confirming the candidate target not is the object occurred after object time, and confirms that the candidate target of the candidate target goes out
Row information cannot generate and first appear object information;It include after object time in the travel time of a candidate target
In the case where time, confirm that the candidate target is the object occurred after object time, and obtain the candidate target in target
History before moment does not occur duration;When history does not occur duration less than target duration, the candidate of the candidate target is confirmed
Object trip information cannot generate and first appear object information.
In the present embodiment, information processing unit is obtained with object time most from the travel time of each candidate target
Closely and earlier than or equal to object time the first history travel time;When the first history travel time being equal to object time, from
In the travel time of each candidate target, obtain with the first history travel time recently and earlier than the first history travel time
Second history travel time, and the duration between the first history travel time and the second history travel time is calculated, obtain history
Do not occur duration;When the first history travel time is earlier than object time, calculate the first history travel time and object time it
Between duration, obtain history and do not occur duration.
In the present embodiment, information processing unit does not occur duration not less than target duration in the history of a candidate target
In the case where, by the candidate target trip information of the candidate target, object information is first appeared as one.
In the present embodiment, information processing unit judges whether a candidate target trip information is real name trip information,
In the case where the candidate target trip information is real name trip information, class object is obtained from each candidate target trip information
Information;It, can not be from the candidate target trip information in the case where the candidate target trip information is not real name trip information
When obtaining real name trip information, confirms that the candidate target trip information cannot generate and first appear object information;Wherein, real name goes out
Row information is the object trip information with class object information.
In the present embodiment, it is being obtained from least one candidate target trip information according to object time and target duration
It takes after first appearing object information, information processing unit obtains at least one and first appear from first appearing in object information
At least one travel time and at least one trip position of object are each first appeared in object;Obtain current time;According to
Preset time difference computation model, at least one travel time and current time, it is corresponding at least to obtain at least one travel time
One time difference;Preset time difference computation model characterize input value and output valve corresponding relationship, input value be the travel time and
Current time, output valve are the time difference;According at least one time difference, determination each goes out for the first time from least one travel time
The newest time of occurrence of existing object and the determining newest appearance position for each first appearing object from least one trip position
It sets.
Information processing unit by time difference the smallest at least one time difference corresponding time of occurrence, as it is each for the first time
There is the newest time of occurrence of object, and by the corresponding appearance position of newest time of occurrence at least one position of going on a journey, as
This first appears the newest appearance position of object.
S104, display first appear object information.
Information processing unit obtains first appear object information after, object information will be first appeared and shown, with
So that first appearing object information according to display, managed to object is first appeared.
Illustratively, information processing unit issues warning information, for example, mentioning according to the trip object information for the first time of display
The side's of warning monitoring first appears object.
In the present embodiment, information processing unit obtain each newest time of occurrence for first appearing object and it is newest go out
Behind existing position, show at least one first appear object, each first appear object newest time of occurrence and it is each for the first time
There is the newest appearance position of object.
Information processing unit also shows the newest time of occurrence for each first appearing object and newest trip position, so that
According to each newest time of occurrence for first appearing object and newest trip position, quickly orient each first appearing object.
Illustratively, as shown in Fig. 2, after inputting target area, object time and target duration, start triggering
In the case where analysis, information processing unit can be handled information, obtain first appearing object information, and go out for the first time
Existing object displayed page shows the name, newest time of occurrence and newest appearance position for first appearing object;Wherein, lead to
After having selected corresponding 132 longitudes and latitudes of 132 video sources to determine target area, object time 2018-07-24, target duration
Be one month, first appearing object is Zhang San, Li Si and king five, the newest time of occurrence of Zhang San be 2018-07-25, it is newest go out
Existing position is Chaoyang District Beijing, and the newest time of occurrence of Li Si is 2018-07-28, newest appearance position is that Beijing Tian An-men is wide
, the newest time of occurrence of king five is 2018-08-03, newest appearance position is Beijing railway station.
It is understood that information processing unit is by presetting at least one object trip information, acquisition is first appeared pair
As, and first appear object for this and generate warning information, object can be first appeared to this and be managed, for example, the police exist
When object time starts to deploy to ensure effective monitoring and control of illegal activities to target area progress emphasis, the personnel for being accused of terrorist activity target before object time is found
It does not appear in target area in duration, but is appeared in target area after object time, issue suspicion terror to the police
The movable biggish prompt of human criminal's possibility.
In the present embodiment, as shown in figure 3, before step S101, information processing method further includes step S301-
S302, as follows:
S301, in the case where reaching at current preset cluster time point, obtain previous default cluster time point and it is current in advance
If clustering acquired image between time point acquires information;
Default cluster time point is preset in information processing unit, is periodically obtained from image collecting device for triggering
Image Acquisition information is taken, and clustering is carried out to Image Acquisition information, for example obtain within one hour or one day simultaneously clustering one
It is secondary.Here, information processing unit is attached with image collecting device, image collecting device real-time image acquisition, preceding default
In the case where clustering time point arrival, information processing unit obtains present image acquisition from image collecting device connected to it
Information.
In the present embodiment, image collecting device may include the device that camera, capture machine etc. are used to obtain image, tool
Body is selected according to the actual situation, and the present embodiment does not do specific restriction.
In the present embodiment, Image Acquisition information includes acquiring image and the corresponding acquisition space time information of acquisition image,
Acquiring space time information includes acquisition time and acquisition position, wherein acquisition position can be according to the position locating for image collecting device
Confidence breath determines, for example, acquisition position can be camera or the corresponding longitude and latitude of capture machine.
S302, according to clustering algorithm, Image Acquisition information and default identity library is preset, obtain presetting at least one object and
Row information;
Information processing unit clusters Image Acquisition information using default clustering algorithm, obtains being drawn according to object
At least one the feature class divided;Obtain at least one corresponding trip information of feature class from Image Acquisition information, and from
The corresponding class identity information of fisrt feature class is determined in default identity library;And then by least one corresponding trip of feature class
Information and the corresponding class identity information of fisrt feature class, composition preset at least one object trip information;Wherein, first is special
Sign class belongs at least one feature class.
In the present embodiment, default identity library is preset in information processing unit, presetting in identity library includes default object
At least one feature class is matched with the default object information that default identity library stores, is obtained by information, information processing unit
Matching result, the process of the acquisition matching result are that the process of real name is carried out to feature class each at least one feature class;Its
In, default object information include it is following at least one: subjects face image, identification card number, gender and social action information, society
Behavioural information is rewards and punishments record of the object in society, for example, being accused of terrorist activity record and previous conviction, previous conviction includes
Criminal type and time of crime etc..
It should be noted that above-mentioned matching result comprises at least one of the following, first is that in default identity library at least one
The all unmatched default object information of feature class, second is that the feature at least one feature class with default object information successful match
Class, third is that at least one feature class with all default object informations all not no feature classes of successful match;It wherein, will at least one
The feature class with default object information successful match in a feature class, as fisrt feature class;It will be matched with fisrt feature class
Successfully default object information, the class object information as fisrt feature class;By being preset with all at least one feature class
The feature class of the non-successful match of object information is as second feature class.
In the present embodiment, presetting identity library may include each identity library divided according to social action information, for example,
There is no the citizenship library of previous conviction, there is the personnel identity library of previous conviction and is accused of the personnel identity library of terrorist activity
Deng.
It should be noted that as shown in figure 4, step S302 is specifically included: step S3021-S3024, as follows:
S3021, an at least identity characteristic image is obtained from acquisition image;
Information processing unit carries out feature identification to acquisition image, filters out the initial pictures with identity characteristic, then right
Initial pictures with identity characteristic are split, and obtain an at least identity characteristic image, at least an identity characteristic image
In every identity characteristic image include an identity characteristic;Wherein, identity characteristic includes face.
S3022, using default clustering algorithm, an at least identity characteristic image is clustered, obtain at least one spy
Levy class;The first similarity at least one feature class between the corresponding identity characteristic image of each feature class is greater than default first
Similarity threshold;
Information processing unit determines biological characteristic from an at least identity characteristic image, using default clustering algorithm to life
Object feature is clustered, at least one feature class is obtained;The first similarity between identity characteristic image can be the same spy
Levy the similarity between any two biological characteristic in class.
Illustratively, as shown in figure 5, each circle is that information processing unit is true from an at least identity characteristic image
The biological characteristic made.
In the present embodiment, identity characteristic image is face-image, and accordingly, biological characteristic is facial biometric;Letter
Processing unit is ceased using default facial biometric recognition methods, and facial biometric is determined from an at least face-image.
Optionally, presetting facial biometric recognition methods includes eigenface method, fish face method, partial binary coding
Histogram etc., is specifically selected according to the actual situation, and the present embodiment does not do specific restriction, and the default face biology is special
Levying recognition methods is existing face recognition method, and details are not described herein for the present embodiment.
In the present embodiment, determined from an at least face-image when information processing unit facial biometric it
Afterwards, information processing unit carries out cluster calculation to facial biometric, and facial biometric is divided according to the first similarity
For multiple feature classes.
It should be noted that information processing unit calculates the first similarity between facial biometric, and by the first phase
It condenses together like the facial biometric that degree is greater than the first default similarity threshold, forms a feature class, a feature
Class is whole facial biometric images to a people.
Illustratively, as shown in figure 5, the similarity value between the dot of identical filling mode is significantly larger than different filling sides
Similarity value between the dot of formula, therefore, the dot of identical filling mode are polymerized to a feature class, obtain 5 feature classes: class
1, class 2, class 3, class 4 and class 5, wherein each feature class indicates an object or a people.
Illustratively, as shown in fig. 6, information processing unit obtains current figure from the corresponding candid photograph library of capture machine (image)
After acquisition information, facial extraction, the face extracted are carried out to every acquisition image in present image acquisition information
Image;It is facial biometric by the biometric features combination of all face-images extracted, and then from facial biometric
It is clustered according to the first default similarity threshold, finally, capture pictures are classified by people, obtains at least one feature class.
S3023, at least one corresponding trip information of feature class is determined from acquisition space time information;
After information processing unit has determined at least one corresponding feature class of present image acquisition information, at least one
Each feature class in a feature class obtains Image Acquisition information corresponding to each facial biometric in each feature class, and
By the acquisition position information and acquisition time information in Image Acquisition information corresponding to each facial biometric, as the spy
The corresponding trip information of class is levied, to also just obtain at least one corresponding trip information of feature class.
S3024, basis preset identity library and at least one feature class, and it is corresponding at least to obtain at least one fisrt feature class
One class object information;At least one fisrt feature class belongs at least one feature class, and default identity inventory contains at least one
Default object information, at least one class object information belong at least one default object information;
Information processing unit presets at least one in object information to be believed with the matched default object of each fisrt feature class
Breath, as the corresponding class object information of each fisrt feature class.
In the present embodiment, it is special to obtain at least one corresponding class center of at least one feature class respectively for information processing unit
Value indicative, at least one fisrt feature class and at least one class central feature value correspond;Calculate at least one default object letter
One group of second similarity of corresponding with current signature class current class central feature value is ceased, in the corresponding current class of current signature class
Heart characteristic value is any one class central feature value at least one class central feature value;Mesh is determined from one group of second similarity
Mark similarity;When target similarity is not less than default second similarity threshold, using current signature class as fisrt feature class, and
From default identity library, the corresponding default object information of target similarity value is determined;Target similarity value is default pair corresponding
Image information is determined as the corresponding class object information of fisrt feature class, and then obtains at least one fisrt feature class corresponding at least one
A class object information.
In the present embodiment, at least one default object information includes at least one subjects face image, information processing apparatus
Set one group of second similarity for calculating at least one subjects face image and the corresponding current class central feature value of current signature class.
In the present embodiment, after information processing unit determines target similarity, it is pre- to judge whether target similarity is greater than
If the second similarity threshold;When target similarity is less than default second similarity threshold, using current signature class as the second spy
Class is levied, i.e. all default object informations in current signature class and default identity library do not have successful match.
Illustratively, information processing unit is similar as target using the maximum similarity of numerical value in one group of second similarity
Degree.
It should be noted that information processing unit is special by class center corresponding to feature class each at least one feature class
Value indicative carries out full dose 1:n with the default object information at least one of default identity library and compares, and takes the second similarity highest and big
This feature class is assigned in the default object information of default second similarity threshold, so that the corresponding people of this feature class is by real name.
S3025, it is corresponded to according at least one corresponding trip information of feature class and at least one fisrt feature class
At least one class object information, obtain presetting at least one object trip information.
Information processing unit determines at least one fisrt feature from least one corresponding trip information of feature class
Corresponding first trip information of class and corresponding second trip information of at least one second feature class, second is special
Sign class belongs at least one feature class;By corresponding first trip information of at least one fisrt feature class and at least one
At least one corresponding class object information of a fisrt feature class is mapped, and real name trip information is obtained;By at least one second
Corresponding second trip information of feature class, as non-real name trip information;By real name trip information and non-real name trip letter
Breath, as presetting at least one object trip information.
Illustratively, information processing unit is special with each first from least one corresponding trip information of feature class
The corresponding trip information of class is levied, as corresponding first trip information of each fisrt feature class.
In the present embodiment, information processing unit is after obtaining presetting at least one object trip information, when getting
When new default object information, new default object information is polymerize with the default object information in default identity library, obtain to
Update default object information;According to default object information to be updated, update is preset at least one object trip information, is updated
Object trip information afterwards.
Illustratively, when entering default identity library in the presence of new default object information, information processing unit is to default identity
Library is updated, for example presets in object information and exist with new default object information with common identity card number when at least one
Default object information, will be in new default object information and at least one default object information with the pre- of common identity card number
If object information merges, and if there is no have phase with new default object information at least one default object information
With the default object information of identification card number, then using new default object information as with it is pre- at least one in default identity library
If the mutually independent default object information of object information, increases newly to default identity library, complete the update to default identity library.Specifically
Ground, information processing unit is after obtaining default monitoring information, when getting new default management object, by new default pipe
Reason object polymerize with the default management object in default identity library, obtains default management object to be updated;According to be updated default
Object is managed, default monitoring information is updated, obtains updated default monitoring information.
In the present embodiment, information processing unit is after obtaining presetting at least one object trip information, next pre-
If cluster time point arrival, collected new figure between current preset cluster time point and next default cluster time point is obtained
As acquisition information;At least one new identity characteristic image is obtained from new images acquisition information;According at least one feature class,
Increment cluster calculation is carried out at least one new identity characteristic image, obtains at least one feature class to be added;From new images
It acquires in information, obtains at least one corresponding new trip information of feature class to be added respectively;It waits adding according at least one
Add the corresponding new trip information of feature class, update presets at least one object trip information, obtains updated object and go out
Row information.
Illustratively, information processing unit needs to acquire information according to new images when collecting new images acquisition information
At least one feature class is updated, to preset at least one object according at least one updated feature class to update
Trip information.Also, in view of the quantity of collected new images acquisition information is larger, information processing unit is adopted according to new images
Collect information and at least one feature class carries out increment cluster, and the trip of at least one object is preset according to the update of increment cluster result
Information.
It should be noted that information processing unit is carrying out increment at least one feature class using new images acquisition information
When cluster, as shown in fig. 7, information processing unit obtains new facial biometric from new images acquisition information, and uses and search
The method at rope class center (for example utilize FAISS index carry out class center search), judge at least one feature class whether there is or not with it is new
Facial biometric corresponding to feature class, obtained corresponding feature class and without two kinds of situations of corresponding feature class, i.e.,
One new facial biometric and the second new facial biometric;Wherein, the first new facial biometric is at least one
There is corresponding feature class in feature class, the second new facial biometric is at least one feature class without corresponding feature class;
The facial biometric new for second obtains new feature class by cluster;Here, by the first new facial biometric and
Second new feature class updates at least one spy as at least one feature class to be added, according at least one feature class to be added
Levy class;Wherein, when updating at least one feature class, the first new facial biometric can be added to it at least one spy
It levies in feature class corresponding in class, the second new feature class is increased newly at least one feature class, obtains at least one more
Feature class after new.
In the present embodiment, information processing unit also utilize at least one feature class to be added at least one feature class into
Row updates, and obtains at least one updated feature class.
Correspondingly, information and/or new default object information pair are acquired according to new images when information processing unit is completed
The update for presetting at least one object trip information, after obtaining updated object trip information, information processing apparatus in S104
It sets according to target area, from presetting at least one object trip information, obtains at least one candidate target trip information, wrap
Include: information processing unit is according to target area, from updated object trip information, obtains the trip of at least one candidate target
Information.
Illustratively, as shown in figure 8, firstly, information processing unit is converted in default identity library at least with identification card number
One default object information, in addition, (present image acquires current in information for collected video flowing or candid photograph image
Acquire image), according to cluster time point triggering cluster was preset at that time, obtain at least one feature class;Then, information processing unit
It at least one is preset into object information and at least one feature class carries out hitting library, obtains three classes archive information, is i.e. foregoing description
Three kinds of matching results, respectively correspond the default object information for no trip information, the fisrt feature class with class object information, with
And the second feature class without class object information;Finally, information processing unit going out based on fisrt feature class and second feature class
Row information first appears object to determine.Wherein, when information processing unit gets new default object information, using new
Default object information updates default identity library, and updates and preset at least one object trip information;When information processing unit obtains
When acquiring information to new images, at least one feature class is updated using new images acquisition information and at least one feature class, and more
At least one object trip information is newly preset, at this point, information processing unit determines for the first time according to updated object trip information
There is object.
It is understood that the trip information of object is characterized due to presetting at least one object trip information, letter
Processing unit is ceased according to target area, object time and the target duration first appeared in inquiry instruction, from presetting at least one
In object trip information, it will be able to get trip information in target area, in the target duration before object time
It is not occurring and what is occurred after object time first appears object information, and then also can be quick, intelligently based on head
Secondary there is object information and determine to first appear object, and the present embodiment improves the feasibility and intelligence searched and first appear object
Property.
Embodiment two
Same inventive concept based on embodiment one, is further detailed.
The present embodiment provides a kind of information processing units 1, as shown in figure 9, the device 1 includes:
Instruction receiving unit 10, for receive first appear inquiry instruction in the case where, looked into response to first appearing
Instruction is ask, obtains target area, object time and target duration in inquiry instruction from first appearing;
First processing units 11, for searching trip information in target area from presetting at least one object trip information
At least one candidate target trip information in domain;
The second processing unit 12, for searching before object time from least one candidate target trip information
It is not occurring in target duration and what is occurred after object time first appears object information;
Display unit 13 first appears object information for showing.
In the present embodiment, presetting each object trip information at least one object trip information includes trip position;
First processing units 11, specifically for the trip position in each object trip information to be not belonging to the trip information of target area
It deletes, obtains at least one candidate target trip information.
In the present embodiment, each candidate target trip information includes each time at least one candidate target trip information
Select the travel time of object;
The second processing unit 12, specifically for the travel time according to object time and each candidate target, judgement is each
Whether candidate target is the object occurred after object time;And occur after object time in each candidate target
In the case where object, the travel time of each candidate target is counted, determines each candidate target before object time
History do not occur duration;And in the case where history does not occur duration not less than target duration, each candidate target is gone out
Row information is used as and first appears object information.
In the present embodiment, first processing units 11 are also used to search from presetting at least one object trip information
Before at least one the candidate target trip information of trip information in target area, target case information is obtained;According to target
Case information determines at least one initial object trip information from presetting at least one object trip information;
Correspondingly, first processing units 11 are specifically used for from least one initial object trip information, search trip letter
Cease at least one candidate target trip information in target area.
In the present embodiment, first processing units 11 are also used to search from presetting at least one object trip information
Before at least one the candidate target trip information of trip information in target area, object screening instruction is obtained;According to object
Screening instruction determines at least one initial object trip information from presetting at least one object trip information;
Correspondingly, first processing units 11 are specifically used for from least one initial object trip information, search trip letter
Cease at least one candidate target trip information in target area.
In the present embodiment, object screening instruction includes that the first object screens instruction and/or the screening instruction of the second object, the
An object screening instruction is for selecting the first class object, and the screening instruction of the second object is for excluding the second class object.
In the present embodiment, the second processing unit 12 are also used to search from least one candidate target trip information
It is occurring in the target duration before object time and after what is occurred after object time first appear object information,
Before display first appears object information, from first appearing in object information, obtains at least one and first appear in object each
First appear at least one travel time and at least one trip position of object;And obtain current time;And according to default
Time difference computation model, at least one travel time and current time, obtain at least one travel time it is corresponding at least one
Time difference;Preset time difference computation model characterizes the corresponding relationship of input value and output valve, and input value is for the travel time and currently
Moment, output valve are the time difference;And according at least one time difference, determination each goes out for the first time from least one travel time
The newest time of occurrence of existing object and the determining newest appearance position for each first appearing object from least one trip position
It sets;
Correspondingly, display unit 13, specifically for showing that at least one first appears object, each first appears object
Newest time of occurrence and the newest appearance position for each first appearing object.
In the present embodiment, device further include:
Information acquisition unit 14, for being obtained from first appearing in inquiry instruction in response to first appearing inquiry instruction
Before target area, object time and target duration, in the case where reaching at current preset cluster time point, obtain previous default
It clusters acquired image between time point and current preset cluster time point and acquires information;And according to default clustering algorithm,
Image Acquisition information and default identity library obtain presetting at least one object trip information.
In the present embodiment, Image Acquisition information includes acquisition image and acquisition space time information;Information acquisition unit 14, tool
Body is used to obtain an at least identity characteristic image from acquisition image;And using default clustering algorithm, to an at least identity
Characteristic image is clustered, at least one feature class is obtained;The corresponding identity characteristic of each feature class at least one feature class
The first similarity between image is greater than default first similarity threshold;And at least one feature is determined from acquisition space time information
The corresponding trip information of class;And according to default identity library and at least one feature class, at least one fisrt feature class is obtained
At least one corresponding class object information;At least one fisrt feature class belongs at least one feature class, presets the storage of identity library
There is at least one default object information, at least one class object information belongs at least one default object information;And according to extremely
Few corresponding trip information of a feature class and corresponding at least one class object letter of at least one fisrt feature class
Breath, obtains presetting at least one object trip information.
In the present embodiment, information acquisition unit 14, it is corresponding at least specifically for obtaining at least one feature class respectively
One class central feature value;And calculate at least one default object information and the corresponding current class central feature value of current signature class
One group of second similarity, the corresponding current class central feature value of current signature class be at least one class central feature value in appoint
One class central feature value;And target similarity is determined from one group of second similarity;And in target similarity not less than default
In the case where second similarity threshold, using current signature class as fisrt feature class, and from default identity library, target phase is determined
Like the corresponding default object information of angle value;And the corresponding default object information of target similarity value is determined as fisrt feature class
Corresponding class object information, and then obtain at least one corresponding class object information of at least one fisrt feature class.
In the present embodiment, information acquisition unit 14 are specifically used for from the corresponding trip letter of at least one feature class
In breath, corresponding first trip information of at least one fisrt feature class and at least one second feature class are determined respectively
Corresponding second trip information, second feature class belong at least one feature class;And it is at least one fisrt feature class is respectively right
At least one the corresponding class object information of the first trip information and at least one fisrt feature class answered is mapped, and is obtained
Real name trip information;And by corresponding second trip information of at least one second feature class, as non-real name trip information;
And by real name trip information and non-real name trip information, as presetting at least one object trip information.
In the present embodiment, device further include: first information updating unit 15, for each according at least one feature class
Self-corresponding trip information and at least one corresponding class object information of at least one fisrt feature class obtain presetting at least
After one object trip information, in the case where getting new default object information, by new default object information and in advance
If the default object information in identity library polymerize, default object information to be updated is obtained;And according to default object information to be updated,
At least one object trip information is preset in update, obtains updated object trip information;
Correspondingly, first processing units 11 are specifically used for from updated object trip information, search trip information and exist
At least one candidate target trip information in target area.
In the present embodiment, device further include: the second information updating unit 16, for each according at least one feature class
Self-corresponding trip information and at least one corresponding class object information of at least one fisrt feature class obtain presetting at least
After one object trip information, in the case where reaching at next default cluster time point, obtains current preset and cluster time point
Collected new images acquire information between next default cluster time point;And at least one is obtained from new images acquisition information
The identity characteristic image of Zhang Xin;And according at least one feature class, it is poly- that increment is carried out at least one new identity characteristic image
Class calculates, and obtains at least one feature class to be added;And from new images acquisition information, at least one spy to be added is obtained respectively
Levy the corresponding new trip information of class;And according at least one corresponding new trip information of feature class to be added, more
At least one object trip information is newly preset, updated object trip information is obtained;
Correspondingly, first processing units 11 are specifically used for from updated object trip information, search trip information and exist
At least one candidate target trip information in target area.
It should be noted that in practical applications, above- mentioned information instruction receiving unit 10, first processing units 11, second
Processing unit 12, display unit 13, information acquisition unit 14, first information updating unit 15 and the second information updating unit 16,
It can be realized by the processor 17 being located on information processing unit 1, specially CPU (Central Processing Unit, center
Processor), MPU (Microprocessor Unit, microprocessor), DSP (Digital Signal Processing, number
Signal processor) or field programmable gate array (FPGA, Field Programmable Gate Array) etc. realize.
The present embodiment additionally provides a kind of information processing unit 1, and as shown in Figure 10, which includes: processor 17, deposits
Reservoir 18 and communication bus 19, memory 18 are communicated by communication bus 19 with processor 17,18 storage of memory processing
One or more executable program of device 17, when one or more program is performed, before being executed such as by processor 17
State any one information processing method described in embodiment.
In practical applications, memory 18 can be volatibility first memory (volatile memory), such as at random
It accesses first memory (Random-Access Memory, RAM);Or non-volatile first memories (non-volatile
Memory), such as read-only first memory (Read-Only Memory, ROM), quick flashing first memory (flash
Memory), hard disk (Hard Disk Drive, HDD) or solid state hard disk (Solid-State Drive, SSD);Or above-mentioned kind
The combination of the first memory of class, and program and data are provided to processor 17.
Present embodiments provide a kind of computer readable storage medium, computer-readable recording medium storage have one or
Multiple programs, one or more program can be executed by one or more processor, when described program is executed by processor 17
Realize any one information processing method as in the foregoing embodiment.
It should be understood by those skilled in the art that, embodiment of the disclosure can provide as method, system or computer program
Product.Therefore, the shape of hardware embodiment, software implementation or embodiment combining software and hardware aspects can be used in the disclosure
Formula.Moreover, the disclosure, which can be used, can use storage in the computer that one or more wherein includes computer usable program code
The form for the computer program product implemented on medium (including but not limited to magnetic disk storage and optical memory etc.).
The disclosure is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present disclosure
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
The above, the only preferred embodiment of the disclosure, are not intended to limit the protection scope of the disclosure.
Claims (10)
1. a kind of information processing method, which is characterized in that the described method includes:
Receive first appear inquiry instruction in the case where, first appear inquiry instruction in response to described, go out for the first time from described
Target area, object time and target duration are obtained in existing inquiry instruction;
From presetting at least one object trip information, it is right to search at least one candidate of trip information in the target area
As trip information;
From at least one described candidate target trip information, search in the target duration before the object time not
It is occurring and what is occurred after the object time first appear object information;
Object information is first appeared described in display.
2. the method according to claim 1, wherein each time at least one described candidate target trip information
Selecting object trip information includes the travel time of each candidate target;It is described from least one described candidate target trip information
In, search head that is not occurring in the target duration before the object time and occurring after the object time
It is secondary object information occur, comprising:
According to the travel time of the object time and each candidate target, judge each candidate target whether be
The object occurred after the object time;
Each candidate target be in the case where the object occurred after the object time, it is right to each candidate
The travel time of elephant is counted, and determines that history of each candidate target before the object time does not occur duration;
In the case where the history does not occur duration not less than the target duration, by each candidate target trip information
Object information is first appeared as described.
3. the method according to claim 1, wherein described from presetting at least one object trip information,
Before searching at least one the candidate target trip information of trip information in the target area, the method also includes:
Obtain target case information;
According to the target case information, at least one initial object is determined at least one object trip information from described preset
Trip information;
Correspondingly, described from presetting at least one object trip information, search trip information in the target area extremely
A few candidate target trip information, comprising:
From at least one described initial object trip information, at least one time of trip information in the target area is searched
Select object trip information.
4. the method according to claim 1, wherein described from least one described candidate target trip information
In, search head that is not occurring in the target duration before the object time and occurring after the object time
It is secondary there is object information after, before first appearing object information described in the display, the method also includes:
It is first appeared in object information from described, obtains at least one and first appear in object and each first appear object at least
One travel time and at least one trip position;
Obtain current time;
According to preset time difference computation model, at least one described travel time and the current time, described at least one is obtained
At least one corresponding time difference of a travel time;The correspondence of preset time difference computation model the characterization input value and output valve
Relationship, the input value are travel time and current time, and the output valve is the time difference;
According at least one described time difference, determination is described from least one described travel time each first appears object
Newest time of occurrence and the determining newest appearance position for each first appearing object from least one described trip position
It sets;
Correspondingly, object information is first appeared described in the display, comprising:
Display it is described at least one first appear object, the newest time of occurrence for each first appearing object and described
Each first appear the newest appearance position of object.
5. the method according to claim 1, wherein first appear inquiry instruction in response to described described, from
It is described to first appear in inquiry instruction before acquisition target area, object time and target duration, the method also includes:
In the case where reaching at current preset cluster time point, obtains previous default cluster time point and the current preset clusters
Acquired image acquires information between time point;
According to default clustering algorithm, described image acquisition information and default identity library, obtains described at least one object of presetting and go out
Row information.
6. according to the method described in claim 5, it is characterized in that, when described image acquisition information includes acquisition image and acquisition
Empty information;The basis presets clustering algorithm, described image acquisition information and default identity library, obtains described presetting at least one
Object trip information, comprising:
An at least identity characteristic image is obtained from the acquisition image;
Using the default clustering algorithm, an at least identity characteristic image is clustered, at least one feature is obtained
Class;The first similarity at least one described feature class between the corresponding identity characteristic image of each feature class is greater than default the
One similarity threshold;
At least one corresponding trip information of feature class is determined from the acquisition space time information;
According to the default identity library and at least one described feature class, at least one fisrt feature class corresponding at least one is obtained
A class object information;At least one described fisrt feature class belongs at least one described feature class, the default identity library storage
There is at least one default object information, at least one described class object information belongs at least one described default object information;
It is corresponding according at least one described corresponding trip information of feature class and at least one described fisrt feature class
At least one class object information obtains described presetting at least one object trip information.
7. according to the method described in claim 6, it is characterized in that, it is described according to the default identity library and it is described at least one
Feature class obtains at least one corresponding class object information of at least one fisrt feature class, comprising:
At least one corresponding class central feature value of at least one described feature class is obtained respectively;
Calculate one group second of at least one described default object information and the corresponding current class central feature value of current signature class
Similarity, the corresponding current class central feature value of the current signature class are any at least one described class central feature value
A class central feature value;
Target similarity is determined from one group of second similarity;
In the case where the target similarity is not less than default second similarity threshold, using the current signature class as first
Feature class, and from the default identity library, determine the corresponding default object information of the target similarity value;
The corresponding default object information of the target similarity value is determined as the corresponding class object information of the fisrt feature class,
And then obtain at least one corresponding class object information of at least one described fisrt feature class.
8. a kind of information processing unit, which is characterized in that described device includes:
Instruction receiving unit, for receive first appear inquiry instruction in the case where, first appear inquiry in response to described
Instruction obtains target area, object time and target duration from described first appear in inquiry instruction;
First processing units, for searching trip information in the target area from presetting at least one object trip information
At least one interior candidate target trip information;
The second processing unit, for searching before the object time from least one described candidate target trip information
The target duration in do not occur and what is occurred after the object time first appear object information;
Display unit described first appears object information for showing.
9. a kind of information processing unit, which is characterized in that described device includes: processor, memory and communication bus, described to deposit
Reservoir is communicated by the communication bus with the processor, and the memory stores executable one of the processor
Or multiple programs pass through the processor and execute claim 1-7 such as and appoint when one or more of programs are performed
Method described in one.
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has program, when
When described program is executed by least one processor, at least one described processor perform claim is caused to require any one of 1-7 institute
The method stated.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910580391.1A CN110309878B (en) | 2019-06-28 | 2019-06-28 | Information processing method and device and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910580391.1A CN110309878B (en) | 2019-06-28 | 2019-06-28 | Information processing method and device and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110309878A true CN110309878A (en) | 2019-10-08 |
CN110309878B CN110309878B (en) | 2022-02-22 |
Family
ID=68077936
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910580391.1A Active CN110309878B (en) | 2019-06-28 | 2019-06-28 | Information processing method and device and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110309878B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111709369A (en) * | 2020-06-17 | 2020-09-25 | 杭州海康威视系统技术有限公司 | Personnel identification method, device, equipment and storage medium |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150365607A1 (en) * | 2013-03-04 | 2015-12-17 | Beijing Lenovo Software Ltd. | Target object information acquisition method and electronic device |
CN106355449A (en) * | 2016-08-31 | 2017-01-25 | 腾讯科技(深圳)有限公司 | User selecting method and device |
CN106709047A (en) * | 2017-01-04 | 2017-05-24 | 浙江宇视科技有限公司 | Object lookup method and device |
CN106778653A (en) * | 2016-12-27 | 2017-05-31 | 北京光年无限科技有限公司 | Towards the exchange method and device based on recognition of face Sample Storehouse of intelligent robot |
CN107783995A (en) * | 2016-08-26 | 2018-03-09 | 杭州海康威视数字技术股份有限公司 | A kind of destination object search method and device |
CN108280489A (en) * | 2018-02-12 | 2018-07-13 | 中国银行股份有限公司 | A kind of sample data generation method and device |
CN109784270A (en) * | 2019-01-11 | 2019-05-21 | 厦门大学嘉庚学院 | A kind of processing method promoting face picture identification integrality |
-
2019
- 2019-06-28 CN CN201910580391.1A patent/CN110309878B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150365607A1 (en) * | 2013-03-04 | 2015-12-17 | Beijing Lenovo Software Ltd. | Target object information acquisition method and electronic device |
CN107783995A (en) * | 2016-08-26 | 2018-03-09 | 杭州海康威视数字技术股份有限公司 | A kind of destination object search method and device |
CN106355449A (en) * | 2016-08-31 | 2017-01-25 | 腾讯科技(深圳)有限公司 | User selecting method and device |
CN106778653A (en) * | 2016-12-27 | 2017-05-31 | 北京光年无限科技有限公司 | Towards the exchange method and device based on recognition of face Sample Storehouse of intelligent robot |
CN106709047A (en) * | 2017-01-04 | 2017-05-24 | 浙江宇视科技有限公司 | Object lookup method and device |
CN108280489A (en) * | 2018-02-12 | 2018-07-13 | 中国银行股份有限公司 | A kind of sample data generation method and device |
CN109784270A (en) * | 2019-01-11 | 2019-05-21 | 厦门大学嘉庚学院 | A kind of processing method promoting face picture identification integrality |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111709369A (en) * | 2020-06-17 | 2020-09-25 | 杭州海康威视系统技术有限公司 | Personnel identification method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN110309878B (en) | 2022-02-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20220092881A1 (en) | Method and apparatus for behavior analysis, electronic apparatus, storage medium, and computer program | |
CN115272037B (en) | Smart city based on Internet of things regional public security management early warning method and system | |
CN106529373A (en) | Human face recognition monitoring alarm method and system | |
TWI747341B (en) | File application method, device and storage medium | |
CN109299642A (en) | Logic based on Identification of Images is deployed to ensure effective monitoring and control of illegal activities early warning system and method | |
CN109784274A (en) | Identify the method trailed and Related product | |
CN110390031A (en) | Information processing method and device, vision facilities and storage medium | |
EP3479270B1 (en) | Incident response analytic maps | |
CN110659391A (en) | Video detection method and device | |
CN110969215A (en) | Clustering method and device, storage medium and electronic device | |
CN112183353A (en) | Image data processing method and device and related equipment | |
CN111291682A (en) | Method and device for determining target object, storage medium and electronic device | |
CN109800674A (en) | A kind of archiving method and device | |
CN109460744A (en) | A kind of video monitoring system based on deep learning | |
CN105608464A (en) | Face recognition mobile terminal solving system and solving method | |
CN108563675A (en) | Electronic record automatic generation method and device based on target body characteristics | |
CN109960969A (en) | The method, apparatus and system that mobile route generates | |
CN112419120A (en) | Group aggregation event early warning method, device and system and electronic equipment | |
CN111291596A (en) | Early warning method and device based on face recognition | |
CN110309878A (en) | A kind of information processing method and device, storage medium | |
CN110321834A (en) | A kind of identity determines method and device, storage medium | |
CN110334232A (en) | Records application method and device, storage medium | |
CN114049658A (en) | Floating population management method and device based on face recognition, computer equipment and storage medium | |
CN109598911B (en) | Early warning method, early warning device and computer readable storage medium | |
CN110490104A (en) | A kind of monitoring system and monitoring method reducing tramp's conflict |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |